当前位置: X-MOL 学术IEEE Micro › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
AsmDB: Understanding and Mitigating Front-End Stalls in Warehouse-Scale Computers
IEEE Micro ( IF 2.8 ) Pub Date : 2020-05-01 , DOI: 10.1109/mm.2020.2986212
Nayana Prasad Nagendra 1 , Grant Ayers 2 , David I. August 1 , Hyoun Kyu Cho 2 , Svilen Kanev 2 , Christos Kozyrakis 3 , Trivikram Krishnamurthy 4 , Heiner Litz 5 , Tipp Moseley 2 , Parthasarathy Ranganathan 2
Affiliation  

It is well known that the datacenters hosting today's cloud services waste a significant number of cycles on front-end stalls. However, prior work has provided little insights about the source of these front-end stalls and how to address them. This work analyzes the cause of instruction cache misses at a fleet-wide scale and proposes a new compiler-driven software code prefetching strategy to reduce instruction caches misses by 90%.

中文翻译:

AsmDB:了解和减轻仓库级计算机中的前端停顿

众所周知,托管当今云服务的数据中心在前端停顿上浪费了大量周期。然而,之前的工作几乎没有提供关于这些前端停顿的来源以及如何解决这些问题的见解。这项工作在整个范围内分析了指令缓存未命中的原因,并提出了一种新的编译器驱动的软件代码预取策略,以将指令缓存未命中减少 90%。
更新日期:2020-05-01
down
wechat
bug